Low-altitude aircraft power supply facility cooperative control method, system, equipment and medium

Abstract

The invention relates to the technical field of data processing, and particularly provides a low-altitude aircraft power supply facility cooperative control method, system and device and a medium, and the method comprises the steps: predicting the renewable energy power generation power of a power supply facility and the charging load power of a low-altitude aircraft based on obtained multi-source information, and generating a prediction scene set representing uncertainty; based on the prediction scene set, through a multi-time scale optimization strategy, solving a scheduling model with optimal operation cost as a target, and successively obtaining a pre-scheduling plan of a day-ahead stage and a real-time control instruction of an intra-day stage; the multi-time scale optimization strategy comprises random optimization based on a scene and feedback optimization based on a rolling time domain; according to the real-time control instruction, power distribution is conducted on the hybrid energy storage system, and a power type energy storage unit preferentially responds to the transient power requirement for charging of the low-altitude aircraft. The comprehensive operation cost is effectively reduced, and particularly, the service life of equipment is prolonged through quantitative energy storage degradation.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data processing, and particularly relates to a low-altitude aircraft power supply facility cooperative control method, system, device and medium. BACKGROUND

[0002] With the rapid development of low-altitude economy, key nodes such as highway service areas are gradually deploying charging facilities for unmanned aerial vehicles, eVTOL and other low-altitude aircraft. However, the existing power supply facilities are mainly designed for ground electric vehicle charging, and its energy system faces severe challenges. First, the charging load of low-altitude aircraft has the characteristics of high power and pulse, which causes a huge impact on the power grid, and the existing power distribution system is difficult to reliably support. Second, the output of distributed photovoltaic and wind power in service areas fluctuates greatly, and the randomness of aircraft load further exacerbates the risk of energy supply and demand imbalance. In addition, the current scheduling strategy fails to fully consider the uncertainty of both the source and the load, and the configuration of the energy storage system is single, which cannot efficiently buffer transient power, resulting in insufficient economic efficiency and reliability of system operation. Therefore, an intelligent method is urgently needed that can cooperatively optimize energy scheduling and flight tasks and realize precise control of hybrid energy storage. SUMMARY

[0003] In view of the above deficiencies of the prior art, the present application provides a low-altitude aircraft power supply facility cooperative control method, system, device and medium to solve the above technical problems.

[0004] In a first aspect, the present application provides a low-altitude aircraft power supply facility cooperative control method, comprising: Based on the obtained multi-source information, the renewable energy power generation of the power supply facility and the charging load power of the low-altitude aircraft are predicted, and a prediction scenario set representing uncertainty is generated; Based on the prediction scenario set, a scheduling model with the optimal operation cost as the target is solved through a multi-time scale optimization strategy, and a pre-scheduling plan for the day-ahead stage and a real-time control instruction for the intra-day stage are obtained in turn; the multi-time scale optimization strategy includes scenario-based stochastic optimization and feedback optimization based on rolling time domain; According to the real-time control instruction, power allocation is performed on the hybrid energy storage system, wherein the power type energy storage unit preferentially responds to the transient power demand of the low-altitude aircraft charging.

[0005] In an optional embodiment, based on the obtained multi-source information, the renewable energy power generation of the power supply facility and the charging load power of the low-altitude aircraft are predicted, and a prediction scenario set representing uncertainty is generated, comprising: Obtain a multi-modal feature sequence composed of historical and real-time data, wherein the multi-modal feature sequence includes traffic flow data, planned low-altitude flight task data, meteorological environment data, calendar time data and historical power data; inputting the multi-modal feature sequence into a pre-trained parallel multi-scale time series prediction model, the model adopting a time series convolution network or a Transformer architecture with fusion attention mechanism to simultaneously output point prediction values and quantile prediction intervals of future renewable power generation and low-altitude aircraft charging load power in a specific time domain; Based on the quantile prediction interval, a scenario generation technique is used to construct a plurality of prediction scenarios and their probability weights for representing joint uncertainty.

[0006] In an optional embodiment, based on the quantile prediction interval, a scenario generation technique is used to construct a plurality of prediction scenarios and their probability weights for representing joint uncertainty, including: Obtaining the quantile prediction interval, determining the marginal probability distribution of renewable power generation and low-altitude aircraft charging load power; Coupling the marginal probability distribution using a Gaussian Copula function to describe the correlation structure between different variable prediction errors and generating a joint scenario set containing correlation features; Each scenario in the joint scenario set contains a complete time series combination and corresponds to a probability weight representing its occurrence probability, and the sum of the probability weights of all scenarios is 1.

[0007] In an optional embodiment, based on the prediction scenario set, a multi-time scale optimization strategy is used to solve a scheduling model with the goal of optimal operation cost, sequentially obtaining a pre-scheduling plan for the day-ahead stage and real-time control instructions for the intra-day stage, including: A stochastic optimization model is established to minimize the expected total operation cost, including the cost of purchasing electricity, the cost of energy storage device degradation, and the cost of wind and light punishment; based on the joint scenario set and its probability weights, the model is solved to obtain a pre-scheduling plan for each day within a complete scheduling period, the pre-scheduling plan including interaction power plan with the main grid and benchmark charging and discharging power plan of the hybrid energy storage system; Entering the actual operation day, taking the second time scale of minute or second as the cycle, based on the updated ultra-short-term prediction data within the second time scale, a model predictive control method is used to solve a finite time domain optimal control problem, and according to the solving result, the benchmark charging and discharging power plan corresponding to the actual operation day is feedback corrected to generate executable control instructions at the current time.

[0008] In an optional embodiment, a stochastic optimization model is established to minimize the expected total operation cost, including the cost of purchasing electricity, the cost of energy storage device degradation, and the cost of wind and light punishment, including:

[0009] wherein, denotes the mathematical expectation, which is calculated by summing over the joint scenario set s ∈ S and its probability weight ; is the total electricity purchase cost to the main grid:

[0010] is the electricity purchase price at time period t, is the power purchased from the main grid at time period t under scenario s; is the equivalent degradation cost of the battery energy storage in the hybrid energy storage system:

[0011] is the degradation cost coefficient, is the charge and discharge power of the battery; is the penalty cost of curtailed wind and solar power:

[0012] is the penalty cost coefficient, is the curtailed renewable energy power at time period t under scenario s; The constraint conditions of the stochastic optimization model include: system power balance constraint, energy storage system operation constraint, power exchange constraint with the main grid, and renewable energy output constraint.

[0013] In an optional embodiment, a model predictive control method is used to rollingly solve an optimal control problem in a finite time domain, and a feedback correction is made to the benchmark charge and discharge power plan corresponding to the actual operation day according to the solving result, including: At the current rolling time k, the benchmark charge and discharge power plan obtained at the current and future time is obtained, and the latest system state measurement value and ultra-short-term prediction information are obtained; A finite time domain optimization problem is constructed, and the objective function includes at least one of the following: a deviation penalty term for tracking the benchmark plan:

[0014] wherein, denotes the summation from i = 0 to H-1, denotes the square of the Euclidean norm, denotes the control input at time k to time k+i, denotes the battery power reference value at time k+i predicted at time k, and H is the prediction time domain length; a smoothness penalty term of the control action:

[0015] wherein, denotes the control input increment from time k to time k+i, and denotes the energy storage power command from time k to time k+i; solving the optimization problem to obtain the optimal control sequence, and issuing the first control command of the optimal control sequence as the corrected actual control command to the energy storage system; repeating the above rolling optimization process at the next sampling time k+1 to realize continuous online feedback correction of the benchmark plan.

[0016] In an optional embodiment, according to the real-time control instruction, the power of the hybrid energy storage system is allocated, wherein the power-type energy storage unit is preferentially responsive to the transient power demand of the low-altitude aircraft charging, comprising: receiving the total power instruction and decomposing it into a high-frequency component and a low-frequency / base load component ; the high-frequency component is preferentially allocated to the power-type energy storage unit to quickly offset the transient power impact caused by the low-altitude aircraft charging; the low-frequency / base load component is allocated to the energy-type energy storage unit to maintain the medium and long-term power balance of the system.

[0017] In a second aspect, the present application provides a low-altitude aircraft power supply facility cooperative control system, comprising: a data acquisition module for predicting the renewable energy power generation of the power supply facility and the low-altitude aircraft charging load power based on the acquired multi-source information, and generating a prediction scenario set representing uncertainty; a plan generation module for solving a scheduling model with the optimal operation cost as the target based on the prediction scenario set through a multi-time scale optimization strategy to obtain a pre-scheduling plan in the day-ahead stage and a real-time control instruction in the intra-day stage in sequence; the multi-time scale optimization strategy includes scenario-based stochastic optimization and rolling time domain-based feedback optimization; a plan execution module for allocating power to a hybrid energy storage system according to the real-time control instruction, wherein the power-type energy storage unit is preferentially responsive to the transient power demand of the low-altitude aircraft charging.

[0018] In a third aspect, an apparatus is provided, comprising: a memory for storing a low-altitude aircraft power supply facility cooperative control program; A processor is configured to implement the steps of the low-altitude aircraft power supply facility collaborative control method as provided in the first aspect when executing the low-altitude aircraft power supply facility collaborative control program.

[0019] Fourthly, a computer-readable medium is provided, on which a low-altitude aircraft power supply facility cooperative control program is stored, wherein when the low-altitude aircraft power supply facility cooperative control program is executed by a processor, the steps of the low-altitude aircraft power supply facility cooperative control method as provided in the first aspect are implemented.

[0020] The beneficial effects of this invention are as follows: the collaborative control method, system, equipment, and medium for low-altitude aircraft power supply facilities provided by this invention significantly improve the system's ability to perceive renewable energy and fluctuating loads through multi-source prediction and uncertainty modeling; by adopting a multi-timescale optimization strategy, it achieves the coordination of day-ahead economic planning and intraday real-time adjustments, effectively reducing overall operating costs, and in particular, extending equipment lifespan by quantifying energy storage degradation; finally, through a hierarchical control strategy of hybrid energy storage, power-type energy storage units prioritize responding to transient power demands, effectively mitigating the impact of low-altitude aircraft charging on the power grid, and greatly improving the reliability and power quality of power supply facilities. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0023] Figure 2 This is a schematic block diagram of a system according to an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0027] The key terms used in this invention will be explained below.

[0028] "Day-ahead phase" and "intra-day phase" are standard professional terms in the field of power system dispatching.

[0029] Day-ahead phase: This refers to the optimized scheduling plan conducted the day before the actual operation date. It formulates a complete daily plan based on information such as load forecasts for the next 24 hours, renewable energy forecasts, and market electricity prices. Its decision-making timescale is typically 15 minutes or 1 hour. Its core function is "planning."

[0030] Intraday Phase: This refers to the rolling adjustment and correction of the day-ahead plan based on the latest system status (such as ultra-short-term forecasts and actual load) on the actual operating day. Its decision-making timescale is shorter, typically 5 minutes, 1 minute, or even seconds. Its core function is "real-time adjustment" or "feedback control."

[0031] The low-altitude aircraft power supply facility collaborative control method provided in this embodiment of the invention is executed by computer equipment, and correspondingly, the low-altitude aircraft power supply facility collaborative control system runs in the computer equipment.

[0032] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention. Wherein, Figure 1 The implementing entity can be a collaborative control system for the power supply facilities of a low-altitude aircraft. Depending on different requirements, the order of the steps in this flowchart can be changed, and some can be omitted.

[0033] like Figure 1 As shown, the method includes: S1. Based on the acquired multi-source information, predict the renewable energy power generation capacity of the power supply facility and the charging load capacity of the low-altitude aircraft, and generate a set of prediction scenarios characterizing the uncertainty; S2. Based on the predicted scenario set, a scheduling model with the goal of optimizing operating costs is solved through a multi-timescale optimization strategy, thereby obtaining the pre-scheduling plan for the day-ahead stage and the real-time control instructions for the intraday stage; the multi-timescale optimization strategy includes scenario-based stochastic optimization and rolling time-domain-based feedback optimization. S3. According to the real-time control command, power is allocated to the hybrid energy storage system, wherein the power-type energy storage unit prioritizes responding to the transient power demand of the low-altitude aircraft charging.

[0034] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0035] S101. Acquisition and Preprocessing of Multimodal Feature Sequences Data sources: Traffic flow data is obtained from highway gantry systems or navigation platform APIs; planned low-altitude flight mission data comes from airspace management or logistics scheduling platforms, including planned take-off and landing times, aircraft type, and energy requirements; meteorological and environmental data is obtained from meteorological bureau APIs, including irradiance, wind speed, and temperature; calendar time data includes time of day and weekday / holiday markers; historical power data is obtained from the local energy management system (EMS) database.

[0036] Data preprocessing: Time alignment of data from different sources (e.g., standardizing to 15-minute intervals), handling missing values, and normalization are performed to form a regular multimodal feature sequence. .

[0037] S102. Training and Inference of Parallel Multi-Scale Temporal Prediction Models Model Structure: The model adopts an encoder-decoder architecture. The encoder uses parallel branches (such as TCN layers with different kernel sizes) to extract multi-scale temporal features, and the decoder adopts a Transformer structure and incorporates an attention mechanism to focus on features that have a significant impact on the prediction target (such as a sudden increase in flight missions).

[0038] Model output: The last layer of the model is designed as a quantile regression layer, using the quantile loss function. The model is trained. It simultaneously outputs point predictions (quantiles of τ=0.5) for the next H time periods (e.g., the next 24 hours) and uncertainty intervals (e.g., quantiles of τ=0.1 and τ=0.9).

[0039] S103. Copula-based Joint Scene Generation Marginal distribution determination: For each predictor variable (such as photovoltaic power) in each future time period t, its quantile prediction results (such as 10th, 50th, 90th) are fitted to an empirical distribution or a parametric distribution of a specific form (such as a Beta distribution) as the marginal probability distribution of the variable in that time period.

[0040] Correlation Coupling: Calculate the historical correlation matrix of the prediction errors of each variable. Using a Gaussian Copula function, connect the marginal distributions of the above variables through an implicit multivariate Gaussian distribution to construct a joint probability distribution that preserves the correlation between variables.

[0041] Scene generation and reduction: A large number of initial scenes (e.g., 1000) are extracted from the joint distribution. Each scene contains a complete time-series curve of all variables across all time periods. Subsequently, scene reduction techniques such as fast forward selection are used to reduce the number of scenes to a computable scale (e.g., 10 representative scenes), and a probability weight π is calculated for each retained scene. s ,satisfy These scenarios and their weights are then used for subsequent stochastic optimization.

[0042] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0043] S201. With the goal of minimizing the total expected operating cost, a stochastic optimization model is established that includes the cost of electricity purchase, the cost of energy storage equipment degradation, and the cost of wind and solar curtailment penalties. Based on the joint scenario set and its probability weights, the model is solved to obtain the daily pre-scheduling plan for a complete scheduling cycle in the future. The pre-scheduling plan includes the power interaction plan with the main grid and the baseline charge and discharge power plan of the hybrid energy storage system. Stochastic optimization models include:

[0044] in, The mathematical expectation is represented by the joint scene set s∈S and its probability weights. Calculate by summation; Total cost of purchasing electricity from the main grid:

[0045] The electricity purchase price for period t. The power purchased from the main network during time period t under scenario s; The equivalent degradation cost of battery energy storage in a hybrid energy storage system:

[0046] This is the degradation cost coefficient. The charging and discharging power of the battery; The cost of penalties for curtailing wind and solar power:

[0047] As a penalty cost coefficient, This refers to the renewable energy power that is abandoned during time period t under scenario s; The constraints of the stochastic optimization model include: system power balance constraints, energy storage system operation constraints, power exchange constraints with the main grid, and renewable energy output constraints.

[0048] System power balance constraint: For each predefined scenario s and each optimization period t, the total power generation of the system must equal the sum of the total load power and losses. The specific expression is:

[0049] in, and These represent the total discharge and charging power of the hybrid energy storage system, respectively. Total load power (including low-altitude aircraft charging load, conventional load, etc.). Let represent the predicted wind power output at time t in the s-th scenario. Let represent the predicted photovoltaic power at time t in the s-th scenario.

[0050] Energy storage system operation constraints: These constraints ensure that energy storage devices operate within safe limits.

[0051] Energy storage state of charge (SOC) constraint: This imposes upper and lower limits on the state of charge of energy storage units (such as batteries), i.e.

[0052] Meanwhile, to ensure scheduling continuity, the SOC at the end of the time period must meet the set value.

[0053] Charging and discharging power constraints: The charging and discharging power of the energy storage unit must not exceed its rated maximum power, i.e.

[0054] Mutual exclusion constraint: At the same time, the energy storage unit cannot be in the charging and discharging state simultaneously. This logical relationship can be modeled by introducing 0-1 integer variables.

[0055] Power exchange constraint with the main grid: Specifies the power exchange limit at the connection point (PCC) between the system and the upper-level main grid, ensuring that it does not exceed the capacity of the line or transformer. This constraint is expressed as:

[0056] in, This represents the maximum allowable exchange capacity (positive values ​​indicate purchasing electricity from the grid, while negative values ​​indicate selling electricity to the grid).

[0057] Renewable energy output constraints: The actual utilization power of renewable energy sources cannot exceed their current maximum available forecast output, and the amount of curtailed solar and wind power should be non-negative. That is:

[0058] in, and This represents the predicted maximum available renewable energy power for time period t under scenario s.

[0059] Solving stochastic optimization models: Discretize the entire scheduling cycle (e.g., the next 24 hours) into T time periods (e.g., 96 15-minute intervals).

[0060] Input the joint scenario set S, where each scenario s contains the power prediction sequence of photovoltaic, wind power, and various loads (low-altitude aircraft, electric vehicles, etc.) for all time periods t and their corresponding probability weights π. s .

[0061] Solve the stochastic optimization model using a commercial optimization solver such as CPLEX, Gurobi, or MATLAB's intlinprog.

[0062] After the solver successfully solves the problem, it outputs the optimal solution, which is the pre-scheduled plan. This plan is deterministic, specifying a concrete operation instruction for each future time period t, but its formulation process fully considers uncertainties (through a set of scenarios).

[0063] Interactive power planning with the main grid: The output is the power value purchased from or sold to the main grid for each time period t. This is a defined planning curve used to trade in the electricity market or arrange transmission routes in advance.

[0064] The baseline charge / discharge power plan for the hybrid energy storage system: The output result is the baseline value of the total charge / discharge power of the hybrid energy storage system for each time period t. This plan serves as a tracking target for the intraday rolling optimization phase, aiming to guide the smooth transition of the energy state (SOC) of energy storage on a macro level to balance intraday fluctuations.

[0065] S202. On the day of actual operation, with a second time scale of minutes or seconds as the period, based on the ultra-short-term forecast data updated within the second time scale, a model predictive control method is used to solve a finite-time domain optimal control problem in a rolling manner, and the reference charging and discharging power plan corresponding to the actual operation day is fed back and corrected according to the solution results, and an executable control command for the current moment is generated.

[0066] 1. Setting the period of the second time scale The "second time scale on the order of minutes or seconds" specifically refers to a fixed period of 1 to 5 minutes. This period is set based on the following engineering considerations: Rapid response requirements: The cycle must be short enough to effectively respond to the power surges of low-altitude aircraft takeoffs and landings, ranging from seconds to minutes.

[0067] Computational complexity: The cycle time needs to be long enough to ensure that the complex calculations of the Model Predictive Control (MPC) optimization problem can be completed within a single cycle. A cycle time of 1-5 minutes provides a reasonable solution time window for edge computing devices.

[0068] In a real system, this periodic parameter can be configured and adjusted in the host computer monitoring system (e.g., preset to 2 minutes). The system clock or high-precision timer is used to trigger the periodic rolling optimization process.

[0069] 2. Update mechanism for ultra-short-term forecast data Update trigger: The execution of the ultra-short-term prediction model is triggered synchronously with the MPC rolling optimization. That is, the ultra-short-term prediction is invoked once at the beginning of each MPC cycle.

[0070] Forecast time domain: The future time domain of the ultra-short-term forecast covers the forecast time domain of MPC, typically from 15 minutes to 1 hour, and uses the same time resolution as the MPC period (e.g., 2 minutes per point).

[0071] Data Input: The input data used by the predictive model is the latest data updated in real time, mainly including: Latest meteorological measurements: Irradiance, wind speed, and other data obtained in real time from meteorological stations deployed in the service area. Latest system status: Including real-time SOC of energy storage and current total load power. Latest mission instructions: Precise aircraft takeoff, landing, and charging reservation information received from the low-altitude flight dispatch system for the next 15 minutes.

[0072] Prediction Execution: Prediction tasks are deployed on edge servers to ensure low latency. The model employs lightweight algorithms (such as ARIMA and online-learned least-squares support vector machines) to complete a prediction computation within seconds, providing the latest prediction data sequence for subsequent MPC optimization. .

[0073] The specific implementation method for using model predictive control (MPC) for rolling optimization and feedback correction during the intraday phase is as follows: 1. Initialization and Data Acquisition On the actual operating day, set the MPC rolling period (e.g., 60 seconds). At each rolling moment k: At the current rolling time k, obtain the current and future baseline charging and discharging power plans, and at the same time obtain the latest system state measurements and ultra-short-term forecast information; Obtain the baseline plan: From the results of the day-ahead optimization phase, read the baseline charge and discharge power plan of the hybrid energy storage system from the current time k to the future predicted time k+H-1, and denot it as the sequence. .

[0074] Acquire real-time data: Obtain the latest system status measurements through sensors and monitoring systems, mainly including the real-time state of charge (SOC(k)) of the energy storage system.

[0075] Get the latest forecast: Call the ultra-short-term forecast module to obtain the updated load and renewable energy generation forecast sequences for the next H steps (covering the forecast time domain). .

[0076] 2. Construction of the Finite-Time Optimization Problem Based on the above information, construct the finite-time optimization problem for the current time k: Decision variables: The optimization variables are the control sequences for the next H steps starting from the current time. , where u represents the charging and discharging power command of the energy storage system to be optimized (positive value for discharging, negative value for charging).

[0077] Objective function: The objective function is designed to minimize tracking deviation and control motion fluctuation, and its specific form is as follows:

[0078] The first item is the tracking deviation penalty, which penalizes the optimized control command u for deviating from the more economical baseline plan. The degree of smoothness; the second item is the control action smoothness penalty item, which penalizes drastic changes in control commands ( Q and R are the corresponding weight matrices.

[0079] Constraints: The optimization problem must satisfy system power balance constraints, upper and lower limits of energy storage SOC constraints, and energy storage charging and discharging power capacity constraints.

[0080] 3. Problem Solving and Command Issuance The constructed optimization problem is solved in real time using a quadratic programming (QP) solver to obtain the optimal control sequence. .

[0081] Using a "rolling time domain" strategy, only the first control instruction in the sequence is used. The command is issued to the power storage converter (PCS) for execution. This command is the actual control command obtained after feedback correction of the original baseline plan, adapted to the latest system state.

[0082] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0083] S301. Total Power Command Reception and Resolution The central controller of the Hybrid Energy Storage System (HESS) receives total power commands from the upper-level Model Predictive Control (MPC) module. (A positive value indicates that the system needs to be discharged, and a negative value indicates that it needs to be charged).

[0084] The power decomposition module uses digital filters (such as first-order low-pass filters or moving average filters) to... Perform real-time frequency domain decomposition. Its transfer function can be simplified to... The time constant τ is set according to the response speed of power storage (usually in the range of milliseconds to seconds) and the main frequency band of the power fluctuation to be smoothed (such as the pulse frequency band corresponding to the charging of low-altitude aircraft).

[0085] The decomposition process is as follows: Low-frequency / base charge components This component changes gradually and mainly includes trend power and background load.

[0086] High frequency components This component includes rapid power fluctuations and impulsive loads, such as the transient power demands of low-altitude aircraft charging.

[0087] S302. Directional distribution of power components High-frequency component allocation: The calculated high-frequency components... Converters that provide real-time and priority power to power storage units (such as supercapacitors or flywheel energy storage). These units have extremely high power density and fast response capabilities (response time down to the millisecond level), and are specifically designed to rapidly absorb or release energy to precisely offset power surges and maintain DC bus voltage stability.

[0088] Low-frequency component allocation: Assigning low-frequency components The converter supplies power to the energy storage unit (such as a lithium iron phosphate battery). This unit is responsible for processing the smoothed base load power, performing hourly energy transfers, and tracking the energy dispatch plan formulated by the upper-level MPC to ensure that its state of charge (SOC) is maintained within the target range.

[0089] S303. Coordination and Protection Logic The central controller continuously monitors the real-time status of each energy storage unit, such as the terminal voltage of the supercapacitor (reflecting its SOC) and the SOC of the lithium battery. When a power-type energy storage unit approaches its energy or power limit, the controller dynamically adjusts the filter parameters or introduces a limiting circuit to transfer some high-frequency components to the energy-type energy storage unit and issues a maintenance warning in the system log to ensure the safe operation of the system.

[0090] In some embodiments, the low-altitude aircraft power supply facility coordinated control system may include multiple functional modules composed of computer program segments. The computer programs for each program segment in the low-altitude aircraft power supply facility coordinated control system may be stored in the memory of a computer device and executed by at least one processor to perform (see details). Figure 1 (Description) Function of coordinated control of power supply facilities for low-altitude aircraft.

[0091] In this embodiment, the low-altitude aircraft power supply facility coordinated control system can be divided into multiple functional modules according to the functions it performs, such as... Figure 2 As shown. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and is stored in memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0092] The data acquisition module is used to predict the renewable energy power generation capacity and low-altitude aircraft charging load capacity of the power supply facility based on the acquired multi-source information, and generate a set of prediction scenarios characterizing uncertainty; The plan generation module is used to solve the scheduling model with the goal of optimizing the operating cost based on the predicted scenario set and through a multi-time-scale optimization strategy, thereby obtaining the pre-scheduling plan for the day-ahead stage and the real-time control instructions for the intraday stage; the multi-time-scale optimization strategy includes scenario-based stochastic optimization and rolling time-domain-based feedback optimization. The planning and execution module is used to allocate power to the hybrid energy storage system according to the real-time control instructions, wherein the power-type energy storage unit prioritizes responding to the transient power demand of the low-altitude aircraft charging.

[0093] Figure 3 The low-altitude aircraft power supply facility cooperative control method provided in the embodiments of this application can be applied to devices. Those skilled in the art will understand that the device structure involved in the embodiments of this invention does not constitute a limitation on the device. A device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the device includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0094] The device 300 may include a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will understand that the server structure shown in the figure does not constitute a limitation of the present invention. It may be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0095] The memory 320 can be used to store execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 is able to perform some or all of the steps in the above method embodiments.

[0096] The processor 310 serves as the control center of the storage device, connecting various parts of the electronic device via various interfaces and lines. It executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 310 may consist only of a central processing unit (CPU). In this embodiment of the invention, the CPU may have a single processing core or include multiple processing cores.

[0097] The communication unit 330 is used to establish a communication channel, enabling the storage device to communicate with other devices. It can receive user data sent by other devices or send user data to other devices.

[0098] The present invention also provides a computer medium, wherein the computer medium may store a program, which, when executed, may include some or all of the steps provided in the embodiments of the present invention. The medium may be a magnetic disk, an optical disk, read-only memory (ROM), or random access memory (RAM), etc.

Claims

1. A method for coordinated control of power supply facilities for low-altitude aircraft, characterized in that, include: Based on the acquired multi-source information, the renewable energy power generation capacity of the power supply facility and the charging load capacity of low-altitude aircraft are predicted, and a set of prediction scenarios characterizing uncertainty is generated. Based on the predicted scenario set, a scheduling model with the goal of optimizing operating costs is solved through a multi-timescale optimization strategy, thereby obtaining the pre-scheduling plan for the day-ahead stage and the real-time control instructions for the intraday stage. The multi-timescale optimization strategy includes scenario-based stochastic optimization and rolling time-domain-based feedback optimization. According to the real-time control command, power is allocated to the hybrid energy storage system, wherein the power-type energy storage unit prioritizes responding to the transient power demand of the low-altitude aircraft charging.

2. The method according to claim 1, characterized in that, Based on the acquired multi-source information, the renewable energy power generation capacity of the power supply facility and the charging load capacity of low-altitude aircraft are predicted, and a set of prediction scenarios characterizing uncertainties is generated, including: A multimodal feature sequence composed of historical and real-time data is obtained, including traffic flow data, planned low-altitude flight mission data, meteorological environment data, calendar time data, and historical power data; The multimodal feature sequence is input into a pre-trained parallel multi-scale temporal prediction model. The model adopts a temporal convolutional network or Transformer architecture with fused attention mechanism to simultaneously output the point prediction value and quantile prediction interval of the renewable energy power generation power and the low-altitude aircraft charging load power in a specific future time domain. Based on the quantile prediction interval, a scenario generation technique is used to construct multiple prediction scenarios and their probability weights to characterize joint uncertainty.

3. The method according to claim 2, characterized in that, Based on the quantile prediction interval, multiple prediction scenarios and their probability weights are constructed using scenario generation techniques to characterize joint uncertainty, including: Obtain the quantile prediction interval to determine the marginal probability distribution of renewable energy power generation and low-altitude aircraft charging load power; A Gaussian Copula function is used to couple the marginal probability distribution to describe the correlation structure between prediction errors of different variables and generate a joint scene set containing correlation features. Each scenario in the joint scenario set contains a complete temporal combination and corresponds to a probability weight representing its probability of occurrence, and the sum of the probability weights of all scenarios is 1.

4. The method according to claim 1, characterized in that, Based on the predicted scenario set, a scheduling model with optimal operating cost is solved using a multi-timescale optimization strategy, resulting in a pre-scheduling plan for the day-ahead phase and real-time control instructions for the intraday phase, including: With the goal of minimizing the total expected operating cost, a stochastic optimization model is established, which includes the cost of electricity purchase, the cost of energy storage equipment degradation, and the cost of wind and solar curtailment penalties. Based on the joint scenario set and its probability weights, the model is solved to obtain the daily pre-scheduling plan for a complete scheduling cycle in the future. The pre-scheduling plan includes the power interaction plan with the main grid and the baseline charge and discharge power plan of the hybrid energy storage system. On the day of actual operation, with a second time scale of minutes or seconds as the period, based on the ultra-short-term forecast data updated within the second time scale, the model predictive control method is used to solve a finite-time domain optimal control problem in a rolling manner. Based on the solution results, the reference charge and discharge power plan corresponding to the actual operation day is fed back and corrected to generate the executable control command for the current moment.

5. The method according to claim 4, characterized in that, To minimize the expected total operating cost, a stochastic optimization model is established, incorporating electricity purchase costs, energy storage equipment degradation costs, and wind and solar curtailment penalty costs. This model includes: in, The mathematical expectation is represented by the joint scene set s∈S and its probability weights. Calculate by summation; Total cost of purchasing electricity from the main grid: The electricity purchase price for period t. The power purchased from the main network during time period t under scenario s; The equivalent degradation cost of battery energy storage in a hybrid energy storage system: This is the degradation cost coefficient. The charging and discharging power of the battery; The cost of penalties for curtailing wind and solar power: As a penalty cost coefficient, This refers to the renewable energy power that is abandoned during time period t under scenario s; The constraints of the stochastic optimization model include: system power balance constraints, energy storage system operation constraints, power exchange constraints with the main grid, and renewable energy output constraints.

6. The method according to claim 4, characterized in that, A model predictive control method is used to solve a finite-time optimal control problem in a rolling manner, and the baseline charge and discharge power plan corresponding to the actual operating day is fed back and corrected based on the solution results, including: At the current rolling time k, obtain the current and future baseline charging and discharging power plans, and at the same time obtain the latest system state measurements and ultra-short-term forecast information; Construct a finite-time optimization problem whose objective function includes at least one of the following: Track deviation penalties for the baseline plan: in, This represents the summation from i=0 to H-1. Represents the square of the Euclidean norm. This represents the control input at time k to time k+i. H represents the battery power reference value predicted at time k+i at time k, where H is the prediction time domain length; Smoothness penalty for controlled actions: in, This represents the control input increment at time k to time k+i, and represents the energy storage power command at time k to time k+i. Solve the optimization problem to obtain the optimal control sequence, and issue the first control command as the corrected actual control command to the energy storage system; At the next sampling time k+1, the above rolling optimization process is repeated to achieve continuous online feedback correction of the baseline plan.

7. The method according to claim 1, characterized in that, According to the real-time control commands, power is allocated to the hybrid energy storage system, wherein power-type energy storage units prioritize responding to the transient power demands of low-altitude aircraft charging, including: Receive total power command And decompose it into high-frequency components. and low-frequency / base charge components ; The high frequency components Priority is given to power-type energy storage units to quickly offset transient power surges caused by charging of low-altitude aircraft; The low-frequency / base charge component It is allocated to energy storage units to maintain the medium- and long-term power balance of the system.

8. A collaborative control system for power supply facilities of low-altitude aircraft, characterized in that, include: The data acquisition module is used to predict the renewable energy power generation capacity and low-altitude aircraft charging load capacity of the power supply facility based on the acquired multi-source information, and generate a set of prediction scenarios characterizing uncertainty; The plan generation module is used to solve the scheduling model with the goal of optimizing the operating cost based on the predicted scenario set and through a multi-time-scale optimization strategy, thereby obtaining the pre-scheduling plan for the day-ahead stage and the real-time control instructions for the intraday stage; the multi-time-scale optimization strategy includes scenario-based stochastic optimization and rolling time-domain-based feedback optimization. The planning and execution module is used to allocate power to the hybrid energy storage system according to the real-time control instructions, wherein the power-type energy storage unit prioritizes responding to the transient power demand of the low-altitude aircraft charging.

9. A collaborative control device for power supply facilities of low-altitude aircraft, characterized in that, include: Memory, used to store the collaborative control program for the power supply facilities of low-altitude aircraft; A processor is configured to implement the steps of the low-altitude aircraft power supply facility collaborative control method as described in any one of claims 1-7 when executing the low-altitude aircraft power supply facility collaborative control program.

10. A computer-readable medium storing a computer program, characterized in that, The readable medium stores a low-altitude aircraft power supply facility collaborative control program, which, when executed by a processor, implements the steps of the low-altitude aircraft power supply facility collaborative control method as described in any one of claims 1-7.

Images